多维数据库模型与应用07(英文).pdf

EE6435 - Lecture 07 Querying Multidimensional Databases (I) EE6435 (Sem A, 2015-16) 1 Conceptual Multidimensional Models • Conceptual Models – Allow better communication between designers and users to understand application requirements – More stable than implementation-oriented (logical) schema, which changes with the platform – Provide better support for visual user interfaces – Recent approaches for conceptual modeling of data warehouses: • Extensions of Entity Relationship (ER) models • Extensions of Unified Modeling Language (UML) • Ad-hoc design models EE6435 (Sem A, 2015-16) 2 Conceptual Multidimensional Models • MultiDim Model: A conceptual multidimensional model – Based on the entity-relationship model – Includes the following concepts in data warehouse and OLAP applications: • dimensions • hierarchies • facts • measures – Supports various kinds of hierarchies existing in real-world applications – Can be mapped to star or snowflake relational structures EE6435 (Sem A, 2015-16) 3 Conceptual Schema of the Northwind Data Warehouse (Fig. 7a) EE6435 (Sem A, 2015-16) 4 Main Components • Schema – It composes of a set of dimensions and a set of facts • Dimension – It composes of either one level or one or more hierarchies • Hierarchy – It composes of a set of levels EE6435 (Sem A, 2015-16) 5 Main Components • Level – Analogous to the entity type in the ER model; E.g., Product and Category are some of